Good advices for a newbie (threading tools lathe)

[PJs]

Thank you very much for the details Stefano. You've clearly done your research!! It makes much more sense to me now and do believe Ulrich's circuit is a good starting place. I also appreciate that you are going the many extra kilometers to eliminate the noise (flame silent) and adjust the frequencies to your needs. A couple of things I don't get about his modulator circuit diagram/designations though.

1. What does TP filter stand for? I get that it is a 1.5MHz filter after the HP filter...basically he has created a band pass between the 2 filters but it seems an odd configuration, almost like a notch filter. Yet the curve shows a clear notch at about 2.6k?? I could be having a brain freeze but don't ever remember one called a TP filter.

2. Why the pull up to 100V feedback to G2 on the EF86. Is that to increase his gain to the ECL82...just seem pretty high? Your coil turns could be why you need more >130V to get it started.


As for the RF side of things a couple come to mind for yours.

First is that Grounding is extremely important as well as shielding. With your transformers on a board and every thing else in the chassis bringing all grounds to "0" is important. A little leakage can be fed into the system and wreak havoc on you grid lines. His shielding also with the perf metal seemed adequate but might consider more of a copper mesh or screen type...more porosity and more wire.

Second is he was very clear that he would build a 3 area system next time for isolation and heat. Again grounding and shielding must be considered strongly but maybe a bit easier to do.

And third the quality of the tube. In your case you have run your tubes to the edge and back several times and may have damaged the cathode/grid connections or warped the plate on yours. Also the filaments may be weakening under the load conditions they've been under. All of that can cause weird things to happen with frequency and capacitance.

The other thing I noticed on the video on FB is it looks like your have many more turns of smaller wire?? Also something about the feed back 1/2 turn and that resistance/capacitance values may need adjusting to temper the feedback to help with your flame control...especially if you are putting 600V into both plates...but may be indicative to damaged tubes?? Also that music/audio on FB was not coming out the plasma right??

It's a fun mystery and sure you will get it. Personally I might roll back to Ulrich's coil/feedback design and try that...if it is still weird check your tubes and grounds/shields. Thanks for the great ride!! ~PJ

[rendoman]

Hi! Thanks for advices!

Got no idea what the second filter at entry of first stage, the first high pass is good to avoid clip of audio in "low" freq, can the second be a Rf filter? I really don't know

I never study ef86 circuit, my audio circuit is a datasheet manual pcl82 amp, with pentode configuration at 272v anode, classic from Philips.
https://frank.pocnet.net/sheets/020/e/EF86.pdf
From this pdf of ef86 rg2 390k, something less Haumann, but pretty close, we can assume that the output voltage is 62v from this stage . I don't know why use a pentode as first stage and a triode of pcl82 as second before pentode modulator... I used triode + pentode for audio and then another pentode, this seems to me more logical, or maybe i'm wrong

I'm trying different flame coil, all with 1mm copper wire, different lenght and diameter, from 20 to 40, 10-40 turns. the only one that run is that on video (calculated as haumann, 8,9uH, if i remember correctly, hope i get in 3 weeks the Lcr meter i buyed from china). 30mm dia like haumann don't run, plus the 100uH on anode is completely unuseful. I used tons of uH more, with a hand made coil on toroid iron powder that help to keep the main coil "safe" . It seems strange, but with 2 coils, the first one near to anode becomes hot really fast.

I need at least 130v to ignite flame in second stage, g2 of pl509 (haumann circuit), that's the voltage applied to g2 of final tube.
you're right with tube quality, I tried change all 3 tubes with high quality new one, but nothing happens. I see another curious thing, I start only filament for 30 sec to warm up, then fire B+, it tooks 2 minutes for reaching the maximum current consumption. Tipical maximum flame with cold tweeter is 150ma, after 2 minutes current raise at 190ma and flame gains body. With this running config I got no problems, soft start, but no same weird things as with Others coils.

Music is not applied to plasma for now, because I want to take one problem at a time
Feedback is not so effective in my test, half or full, position in middle of tungsten - coil - up tip gives few results, even the 10nF , I changed with 6,8 3 and 1 nF but I didn't see something different.
Resistor G1 - loop changes polarisation of tube, this make differences.
g2 voltage in my setup is higher than Haumann, I can't fire till 130v, for sure it's a question of different tube, russian tube have poor datasheet , but i can maybe use data from LS50 german tube, the gu50 should be a copy of this

http://www.mif.pg.gda.pl/homepages/f...043/l/LS50.pdf

Next days I will make something for frame

[PJs]

Thanks Stefano. Food for thought from experience...When I get to a stuck point it is easy to get into a hurry up, try this try that mode and usually forget to document what I did and the Results. I relate it to trying to fight my way out of the inside of a balloon! Usually I find that letting it go for a day or two, something bubbles to the surface that either puts me on the right track or the actual solution. Allow some Ω time. ~¿@

The 1.5Mhz filter is a bit odd to me too, but it's mainly the configuration of the 2 filters in series. I do not consider it (1.5Mhz) an RF filter in this circuit.

As for the coil builds...Start with the calcs and work a little either side of what is needed. The 10-40 turns is a big swing...especially changing diameters too. Document each!!

Had another thought about the coil and the proximal capacitance effects. Maybe you are experiencing issues because of that??

As they say...Rome wasn't built in a day...but it did get built...as well as the Sistine chapel. ~¿@ You'll get it built too! Thanks for sharing with us. ~PJ

[rendoman]

Hi!
Big update tonight
The new coil is running better than others! Full gas flame it's a little bigger now, but the main thing is that now lowest flame possibile (with closed potentiometer) is stable and is +/- 6mm after warmup. Tube oscillates better, less heat till 150ma, after grid become red, but it's good, in this conditions it's 70v over maximum rating. I need the measurement tools to drop a line, but I'm happy to see that I'm on the right side now, even if there's a lot of work to do! An other good thing is that power supply make less noise and coils runs colder.

Got a question for you, I noticed a change in flame color, till 1cm +/- it's a bright purple, when I push the limit flame becomes white, not only in the flake (idk the correct name, it's the outside of flame) but even the dart is clearer. Actually I don't know if it's matter of power, or maybe some frequency change. Next days I will try to vary the tuning freq, Hope measurement instruments will arrive soon! I need to know at which frequency the oscillator is running.

making of the new coil


play with flame, the last picture is the minimum new flame at full closed potentiometer





Olive all'ascolana and strawberries kiwi cream pie
Ricetta Olive all' ascolana - Le Ricette di GialloZafferano.it
https://www.finedininglovers.com/sto...ives-ascolana/

[PJs]

Looks like you are getting a pretty good corona now, small and tight! Nice Going!! Good your power supply and coils aren't straining (humming) now. Hopefully your new instruments will help answer some questions for you about frequency/power etc.

Not sure about your question but think you are questioning/witnessing the oxidation of the atmosphere with the larger white flame/corona from more power being applied (picture 6)? It may also be some oxidation of your needle as the picture has some yellowing out toward the edge. Forgot what material but seem to remember tungsten alloy for your tip??

Question for you. I can't tell from the pictures but have you moved the 3/4T coil up and down slightly?

Thanks for the update, nice progress. ~PJ

[rendoman]

Hi!
you're right, the tungsten rod shows some oxidation, I can try grinding the tip in order to see if something change, maybe the flame can oxidize the tip continuously during operation? I remember old chemical exp with copper sulfate and flame, it can be something similar in fact

I tried different position and different capacitor + resistor of the 3\4 reaction loop. separate discussion for components, because they change with main coil, i adjust the position of this loop with neon tube. There's a range of half inch about in which the strenght of the field is stronger, using neon tube it's pretty easy to locate the best position

I see another good thing with the new tuning, semi faraday gage is still helpful, but the negative interaction with the environment is decreased, as temperature of coils and even temp of power transformer. For sure it drains less current relating flame size. Excepts for overdrive, datasheet of tube said maximum operation value of 250v, at this driving voltage i got 130ma at flame, with great size.
I'm going to make some modifications on circuit, hope to got some news in few hours

[rendoman]

Hi!
This evening I tried some experiments with audio, for sure strange things happened! Got huge troubles with some audio tube connection, maybe due to oxidation of pcl82 socket, then terrible Gnd hum and buzz from flame. I tried different things, the big problem was m6 inox allen plus nut close to reaction ring, it's a couple of week that I see strange behavior of that component. I remove the part and the flame starts oscillating good with no hum sound. Impressive? This circuit is really bastard inside!
Few things about video, the cam is a china gopro, sound is not good, audio tape is really old and walkman has poor output (i tried with my Lm3886 amp of 15"sub , very low output). There is a lot of work to do, sound is pretty good! i'm not sure about xover filter at imput of plasma, i think I'll remove soon. Volume is not so bad with 110\120ma flame, more or less gasoline gives few changes in spl, I found the best position with this value. Got some bad rumor about audio first stage, pentode loading resistor 20w ceramic type runs pretty high temperature during operation, it's barely touchable with finger (I got trained welding fingers ), imho this huge amount of heat can be avoid using a output transformer, maybe some more power can be achieved! It's early to speak about, but my impression is that amplifier is not enough, maybe it needs more then a small pcl82 pentode to run well, or more likely I have to optimize the final circuit and then think to audio part. The good news is that the coils are cold, but the B+ anode transformer after 3 hours of game is something about 60° , like a normal tube amp +/-
Ps: the cable is audiophile grade! You don't beleve me? Maybe you're right! at least welds are good

[PJs]

Congratulations Stefano, you have successfully modulated plasma with audio!! A big feat! Couple of observations/questions.

Did you feed your tape directly to the grid of the tube? What is the output level of the tape in volts?

There was a definite clipping at several points in the video. Not sure what you were doing at the time with the controls but a few of them felt/sounded like a low frequency ground pounding clip...Ground noise build up through a cap?

Do you have any way to feed a frequency generator sweep into it with a constant output voltage?

Baby steps now...to dial it in.

Terrific! ~PJ

[rendoman]

Many thanks!
I havn't measured the output voltage of walkman, but I want to use a soundcard of old pc to make some test, with this I can use even a frequency generator. At this time i don't want to use my Scarlett 2i2 soundcard, I prefer be safe
The tape is applied to a high pass filter at imput (don't know if it's running good or not) then triode + pentode of Pcl82 tube. Flame now is better but there's lot of work to do, it's not stable with audio connected. Troubles with starting and restarting, it's a clear sign of bad tuning frequency of coils imho. In video I try different position of both volume control and flame:
- sound is nice, clear and loud at low-mid flame. optimum is from minimum 60 ma to max 100\110ma, when big flame appears volume lowered and quality decreases.
- if I raise the volume of audio too much, the flame turns off, i think the oscillator stop running, and it's difficult to restart

I think the trouble is the oscillator and the length of cables - position of components. Maybe 3w pentode audio stage is not enough to play louder, there's still this possibility
I'm focusing on main and block coils, I found that they are responsible for the major changes, I just finished a new double turns coil, as expected circuit runs better. Hope freq counter will arrive in few days, that's my only way to see at which freq circuit is running.

you're right! now baby step each time, it's like "labor limae"
I'm thinking of close all side of plasma, put in orizontal position and regroup in small area all mammut for changing components, not an easy job, but maybe the best choice

[rendoman]

sorry I forgot pictures